1. Technical Field
The present invention relates to computer network environments and more specifically to non-visual presentation of electronic documents.
2. Description of Related Art
Information on the World Wide Web is typically made available by structuring the information into a visual presentation. Hyper Text Markup Language (HTML) is used by web authors to define the visual structure. The end user is presented with this information by viewing the information on a computer display, after the information has been rendered into a visual format by a web browser (e.g. Netscape Navigator or MS Internet Explorer).
Web sites of well established businesses and organizations make extensive use of visual images. A HTML MAP defines a set of sub-regions over the image area. Each region is called an AREA, and is defined by an AREA element within the MAP definition. Each AREA can be associated with an Internet Uniform Resource Locator (URL). When the end user performs a mouse click within an area defined by the MAP, the web browser will navigate the associated URL. This process works well for a sighted user who is accessing the web using a visual browser. However, this process is not accessible by people with vision impairments, nor is it accessible by users who do not have a visual display device available (e.g. while driving a car).
A variety of software products are becoming available which enable non-visual access to HTML pages. These products capture the web page content and then present an audible rendering of the web page. This is generally accomplished by using a text-to-speech (TTS) technology to read the textual content.
HTML, which is used to provide a visual structure to a web page, also provides a semantic structure to the page. Well known techniques exist for parsing an HTML source file into a parse tree, also known as a Document Object Model (DOM). The various structural elements and relationships among the elements are then apparent from the topology of the parse tree. The DOM is accessible as a component, and this component provides the foundation needed to build a non-visual browser.
In an HTML page, a MAP-AREA definition is a non-visible element. Consequently, the web author is free to locate the MAP-AREA definition within the DOM wherever the author pleases. A cross referencing scheme is then used to associate the MAP-AREA definition with a corresponding IMAGE within the DOM. However, the physical separation of the IMAGE from the MAP-AREA definition introduces a fair amount of program complexity when the HTML page is being presented by a non-visual browser. Currently, non-visual browsers must maintain extensive internal records in order to keep track of the logical association between a MAP-AREA and an IMAGE in the DOM.
Therefore, it would be desirable to have a method for maintaining the logical association between corresponding IMAGE-AREAs and IMAGEs in a DOM, while reducing program complexity and the need for extensive record keeping.